“This position paper argues that classic systems security design principles are what we need to secure AI agents.”
Agent Security is a Systems Problem
We take the position that agent security must be approached as a systems problem: the AI model powering the agent must be treated as an untrusted component, and security invariants must be enforced at the system level. Through this lens, efforts to increase model robustness (the dominant viewpoint in the community) are insufficient on their own. Instead, we must complement existing efforts with techniques from the systems security domain. Based on our experience as cybersecurity researchers in operating systems, networks, formal methods, and adversarial machine learning, we articulate a set of core principles, grounded in decades of systems security research, that provide a foundation for designing agentic systems with predictable guarantees. As evidence, we analyze eleven representative real-world attacks on agents and discuss how systems principles, if realized, could have prevented these attacks. We also identify the research challenges that stand in the way of implementing these principles in agents.
候选层:至少 2 个不同的人、在 2 个不同月份提到过,还没到主书架的 3 人 3 个月。没有做立场标注,所以只写“被提及”。
HN 书评
没有评论直接链到它;下面是它被提交的讨论串里回复最多的评论(不计入统计)“agent security is a very old lesson UNlearned. data and instructions don't mix. And agents can't separate them”
- Agent Security Is a Systems Problem ↗sbulaev · 2026-05-20 · 1 分
- Agent Security Is a Systems Problem ↗yakkomajuri · 2026-05-28 · 2 分
- Agent Security Is a Systems Problem ↗bradreaves2 · 2026-07-17 · 2 分